The invention relates to a method for measuring temporally constant signal ratios, in particular ionic currents for measuring isotope ratios, preferably by using a mass spectrometer, two or more signals being measured in parallel, that is to say being caught by a catching device having two or more catchers, and evaluated in a measuring device having two or more channels. The invention also relates to a catching device and a mass spectrometer having such a catching device.
The invention proceeds from a prior art as disclosed in DE 31 39 975 C2 (corresponding to U.S. Pat. No. 4 495 413). Reference is expressly made to the prior art described there.
The main field of application of the invention is the high precision measurement of isotope ratios. Mass spectrometers having a plurality of ion catchers are known to be used for this purpose. The separate ionic currents are measured in parallel in these catchers. As a result, the desired measurements can be carried out substantially more quickly and precisely than in mass spectrometers having only one catcher. Temporal fluctuations in the signal intensities occur at all catchers simultaneously and do not influence the measuring accuracy for the signal ratio.
Each catcher has a dedicated electronic measuring system and is part of a measuring channel. Constituents of the electronic measuring systems are, for example, current amplifiers, voltage-to-frequency converters, digital voltmeters or digital ammeters. The ionic currents caught by a catcher are measured (amplified) and evaluated in the assigned electronic measuring system.
Being able to intercompare the measured signals quantitatively requires calibration of the electronic measuring systems of the various measuring channels relative to one another. It is precisely this which is done by the method described in the Patent ""431 specified above. The gains Gi thus determined for the individual signal amplifiers (current amplifiers) i can, however, be measured in practice only with a limited accuracy dGi. The accuracy dGi is given in practice by the noise of the amplifiers and the noise of the calibration source, and by the limited stability of the overall system. The relative calibration uncertainty dGi/Gi in the best present-day systems is approximately equal to 5 ppm (1 Stdev) per measuring channel. This calibration uncertainty represents the limit of the measuring accuracy with present-day systems.
The isotope ratio IR(A,B) of the isotopes A and B is calculated as follows taking account of the gains G1 and G2:                               IR          ⁢                      xe2x80x83                    ⁢                      (                          A              ,              B                        )                          =                                            G              1                        xc3x97                          I              A                                                          G              2                        xc3x97                          I              B                                                          Equ.  1            
A calibration uncertainty dG1 is continued directly into the uncertainty of the measured isotope ratio IR(A,B) as:                                           ⅆ            IR                    ⁢                      xe2x80x83                    ⁢                      (                          A              ,              B                        )                          =                                            I              A                                                      G                2                            xc3x97                              I                B                                              xc3x97                      ⅆ                          xe2x80x83                        ⁢                          G              1                                ⁢                      xe2x80x83                    ⁢          and                                    Equ        .                  xe2x80x83                ⁢        2                                                                    ⅆ                              xe2x80x83                            ⁢              IR                        ⁢                          xe2x80x83                        ⁢                          (                              A                ,                B                            )                                            IR            ⁢                          xe2x80x83                        ⁢                          (                              A                ,                B                            )                                      =                              ⅆ                          xe2x80x83                        ⁢                          G              1                                            G            1                                              Equ        .                  xe2x80x83                ⁢        3            
Since the calibration of the two amplifiers involved (gains G1, G2) are mutually independent, the uncertainty defined in equation 3 can be estimated as follows:                                                         ⅆ              IR                        ⁢                          xe2x80x83                        ⁢                          (                              A                ,                B                            )                                            IR            ⁢                          xe2x80x83                        ⁢                          (                              A                ,                B                            )                                      =                                                            (                                                      ⅆ                                          xe2x80x83                                        ⁢                                          G                      1                                                                            G                    1                                                  )                            2                        +                                          (                                                      ⅆ                                          xe2x80x83                                        ⁢                                          G                      2                                                                            G                    2                                                  )                            2                                                          Equ.  4            
With a relative uncertainty of                                           ⅆ            G                    G                ≅                  5          ⁢                      xe2x80x83                    ⁢          ppm                                    Equ.  5            
per measuring channel, the overall result of this is a measuring uncertainty of                                                         ⅆ              IR                        ⁢                          xe2x80x83                        ⁢                          (                              A                ,                B                            )                                            IR            ⁢                          xe2x80x83                        ⁢                          (                              A                ,                B                            )                                      ≅                  7          ⁢                      xe2x80x83                    ⁢          ppm                                    Equ.  6            
This precision represents the limit of present-day technology.
In fact, the calibration uncertainty can also be improved by frequent repetition. With a calibration time per measuring channel of approximately one minute, approximately ten minutes result as overall calibration time with the in practice frequently up to ten measuring channels. The measuring system would have to be calibrated approximately 25 times for the statistical improvement of the reliability of the calibration from approximately 5ppm to 1 ppm. The mean value of the calibrations would then have a precision of 1 ppm, assuming the overall system has remained stable over the time interval of 250 minutes at 1 ppm. In practice, this condition can be ensured only with difficulty.
It is the object of the present invention to create a method with the aid of which the measuring uncertainty can be further reduced overall. In particular, the starting point in this case is the method described in the above patent.
According to the invention, the object is achieved by means of the following features:
a) the measurement is subdivided into sequential time periods (measuring blocks),
b) in a first time period a specific catcher is assigned to each electronic measuring system and
c) the assignment of the catcher and electronic measuring system is changed in the next time period so that then each electronic measuring system is assigned to a different catcher than in the preceding time period.
The measurement results thus obtained are subjected to averaging. The influence of the calibration uncertainty of the individual measuring channels on the achievable precision is thereby eliminated. The method is advantageous, in particular, for measuring ionic currents when determining isotope ratios. Moreover, it is possible using the invention generally to improve the measurement of temporally constant signal ratios. The term xe2x80x9ccatcherxe2x80x9d then relates to any device which catches or records signals.
Advantageously, measurements are carried out over at least as many time periods as there are measuring channels, that is to say electronic measuring systems with catchers, present. Each catcher is thereby interconnected at least once with each electronic measuring system.
To carry out the method, the measuring system used is characterized by a device (relay matrix) for alternately interconnecting the catchers and the electronic measuring systems in such a way that each electronic measuring system can alternately be connected to one of a plurality of catchers. The relay matrix is thus a switch system for the alternating connection of the catchers to the electronic measuring systems.
Further features of the invention follow from the remainder of the description and from the claims. In this case, the invention also comprises a mass spectrometer having a measuring system.